https://github.com/pytorch/TensorRT/pull/2859 PR uses torch.empty which initializes the tensor on CPU and hence we get device location errors. We can initialize to our default device which might fix this.
To Reproduce
Steps to reproduce the behavior:
1.
2.
3.
Expected behavior
Environment
Build information about Torch-TensorRT can be found by turning on debug messages
Torch-TensorRT Version (e.g. 1.0.0):
PyTorch Version (e.g. 1.0):
CPU Architecture:
OS (e.g., Linux):
How you installed PyTorch (conda, pip, libtorch, source):
Build command you used (if compiling from source):
Are you using local sources or building from archives:
Bug Description
https://github.com/pytorch/TensorRT/pull/2859 PR uses torch.empty which initializes the tensor on CPU and hence we get device location errors. We can initialize to our default device which might fix this.
To Reproduce
Steps to reproduce the behavior:
1. 2. 3.
Expected behavior
Environment
conda
,pip
,libtorch
, source):Additional context